Most of the advice on this is written by agencies, which makes it worth reading sceptically — including this. Here are the questions that actually separate outcomes, and the answers that should worry you.
Ask who writes the code
The question: will the people who scoped this be the people who build it?
The answer that should worry you: an evasive one. The bait-and-switch — senior people in the pitch, juniors on the project — is the most common failure mode in agency work, and it never appears in the contract.
Ask what happens to the code
The question: who owns the source, the pipelines and the store accounts?
The answer that should worry you: anything other than "you do, on payment, and it's in the agreement." Publishing your app under an agency's developer account is a hostage situation with a friendly name.
Ask how change requests are priced
Scope will change. The question is whether repricing happens in the open or arrives as a surprise invoice in month four.
Ask for the reasoning, not the recommendation
Any competent team can recommend a stack. Ask why — and ask what would change their mind. A platform recommendation that doesn't come with the conditions under which it's wrong is a preference, not an argument.
Ask what they've turned down
An agency that has never declined a project is optimising for utilisation, not for your outcome.
Check the things they can't control
- Real references you can contact, not logos on a page
- Whether their own apps are still maintained
- Third-party profiles — Clutch, GoodFirms — where reviews aren't curated
What a good scope looks like
Written. Specific about what's included and what isn't. Honest about uncertainty. A number the team can stand behind beats a number that wins the pitch, and you can usually tell the difference by reading it.
Our answers
They're on the FAQ and the about page, and the pricing is published rather than gated on the cost page.